Echocardiography Underestimates Stroke Volume and Aortic Valve Area: Implications for Patients With Small-Area Low-Gradient Aortic Stenosis Calvin W.L. Chin, MD, Hwan J. Khaw, BSc, Elton Luo, BSc, Shuwei Tan, BSc, Audrey C. White, CRCS-AE, David E. Newby, MD, PhD, Marc R. Dweck, MD, PhD Canadian Journal of Cardiology Volume 30, Issue 9, Pages 1064-1072 (September 2014) DOI: 10.1016/j.cjca.2014.04.021 Copyright © 2014 Canadian Cardiovascular Society Terms and Conditions
Figure 1 Estimation of left ventricular outflow tract (LVOT) area using echocardiography and magnetic resonance imaging. (A) The LVOT diameter was measured at the aortic cusp insertion points (red arrows) in the parasternal long axis view. The LVOT area was estimated from the diameter measured. (B) The stroke volume was calculated as the difference between end-diastolic and end-systolic volumes. Planimetry of the endocardial borders (red contours in end-diastolic and end-systolic frames) was performed including the papillary muscles and minor trabeculations in volume measurements during both phases of the cardiac cycle. Left ventricular mass was calculated by multiplying the total end-diastolic myocardial volume (green and red contours in the end-diastolic frame) by the specific gravity of the myocardium (1.05 g/mL). Papillary muscles and minor trabeculations were excluded in mass measurements, with care taken to avoid right ventricular trabeculations. (C) Planimetry of the LVOT area in the coaxial short axis view on cardiovascular magnetic resonance imaging at mid-systole. Canadian Journal of Cardiology 2014 30, 1064-1072DOI: (10.1016/j.cjca.2014.04.021) Copyright © 2014 Canadian Cardiovascular Society Terms and Conditions
Figure 2 Stroke volume correlation and Bland-Altman analysis. Doppler stroke volume correlated weakly with magnetic resonance imaging (MRI) stroke volume (A), with a fixed bias and wide limits of agreement (B). In 40 patients, stroke volume was calculated using planimetered left ventricular outflow tract area on MRI and Doppler left ventricular outflow tract flow (MRI-Doppler). This approach demonstrated excellent correlation with MRI stroke volume (C), without significant bias (D). Canadian Journal of Cardiology 2014 30, 1064-1072DOI: (10.1016/j.cjca.2014.04.021) Copyright © 2014 Canadian Cardiovascular Society Terms and Conditions
Figure 3 Left ventricular outflow tract (LVOT) area correlation and Bland-Altman analysis. Although LVOT area estimated using echocardiography demonstrated a moderate correlation with planimetered LVOT area on magnetic resonance imaging (A), the echocardiographic LVOT area underestimated the planimetered area with wide limits of agreement (B). Canadian Journal of Cardiology 2014 30, 1064-1072DOI: (10.1016/j.cjca.2014.04.021) Copyright © 2014 Canadian Cardiovascular Society Terms and Conditions
Figure 4 Aortic valve area corrleation and Bland-Altman analysis. Aortic valve area estimated using Doppler stroke volume and magnetic resonance imaging-derived stroke volume demonstrated poor agreement and significant underestimation (A), despite excellent correlation (B). Canadian Journal of Cardiology 2014 30, 1064-1072DOI: (10.1016/j.cjca.2014.04.021) Copyright © 2014 Canadian Cardiovascular Society Terms and Conditions
Figure 5 Relationship between aortic valve area and mean pressure gradient. The aortic valve area was calculated from the continuity equation using Doppler stroke volume. An aortic valve area of 1.0 cm2 corresponded to a mean pressure gradient of 24 mm Hg (A). Correcting these values using the magnetic resonance imaging stroke volume, an aortic valve area of 1.0 cm2 corresponded to a mean pressure gradient of 37 mm Hg (B). Canadian Journal of Cardiology 2014 30, 1064-1072DOI: (10.1016/j.cjca.2014.04.021) Copyright © 2014 Canadian Cardiovascular Society Terms and Conditions
Figure 6 Reclassification of aortic stenosis severity. Using traditional echocardiographic measurements and the recommended severity cutoffs established in current guidelines (A), 56 patients had discordant small-area low-gradient aortic stenosis. Twenty patients were reclassified to concordant nonsevere aortic stenosis when cardiovascular magnetic resonance imaging stroke volume was used to estimate aortic valve area (B). A further 7 patients were reclassified as having concordant severe disease using the revised thresholds of 1.0 cm2 and 37 mm Hg (C). The corresponding pie charts show the flow states in patients with discordant small-area low-gradient aortic stenosis (stroke volume estimated using cardiovascular magnetic resonance imaging). Canadian Journal of Cardiology 2014 30, 1064-1072DOI: (10.1016/j.cjca.2014.04.021) Copyright © 2014 Canadian Cardiovascular Society Terms and Conditions